Feature Requests (132)

All issues labeled as feature requests. Sourced from the official GitHub repo; status derived from official labels; every entry links back to its source with a last-checked date.

#58340 Investigating P3 comp/agent

[Feature Request] Auto-load relevant skills before task execution

Feature request: auto-suggest or auto-load relevant skills before task execution, since agents often forget to check installed skills.

#58325 Investigating P3 comp/agent

Integration proposal: Hermes as a personal agent in a governed Gotong workflow/life-flow hub

Integration proposal from the Gotong project: Hermes as the personal agent layer inside a governed workflow hub.

#58301 Fixed P3 comp/dashboard

Native iOS sessions list infinite scroll should load older pages automatically

The native iOS sessions list relied on a manual 'Load older sessions' button instead of auto-loading on scroll; this has been fixed to trigger pagination automatically with duplicate-load guards.

#58300 Fixed P3 comp/dashboard

Native iOS session filters should show total counts per source

The native iOS app's session filter chips and section headers derived counts from already-loaded (paginated) sessions rather than full backend totals, which was misleading; this has been fixed to use full totals per source.

#58274 Investigating P3 comp/cron

[Feature]: Full Profile Parity — Dashboard, Sessions, Cron, and Path Isolation

Umbrella feature request describing multiple subsystems (cron, dashboard, desktop) that don't properly support non-default Hermes profiles, since the codebase assumes ~/.hermes/ is the canonical root instead of consistently resolving ~/.hermes/profiles/<name>/.

#58248 Investigating P3 comp/gateway

Delivered-text persistence surfaces bypass PII/redaction hooks (rich-sent index, delivery mirror, API error dumps)

Hermes 0.18 persists delivered (post-redaction-restore) conversation text to disk in multiple surfaces (including rich_sent_index.json and the delivery mirror) without any redaction hook or TTL, which is a problem for deployments using PII tokenization plugins that restore PII only at delivery time.

#58217 Investigating P3 comp/agent

[Feature]: Give pre_llm_call a last_turn record — let plugins perceive what the previous turn did

The pre_llm_call plugin hook can only see what a turn is about to do, not what the previous turn actually did, since that state is wiped at the start of the next turn; the issue proposes a last_turn record passed into the hook.

#58198 Investigating P3 comp/cli

[Feature]: The selected Model and provider label should appear at the top of the list for easy access

Users have to repeatedly scroll to find the currently selected model/provider in the picker list; the request asks for the selected entry to always appear at the top.

School diagnosis & fix
#58197 Workaround P2 comp/agent

[Feature]: Retry malformed tool calls with forced tool_choice

Local/quantized models intermittently emit malformed tool calls that get silently replaced with empty args or re-fail on retry; the proposal pins tool_choice to the failed tool on retry, already implemented per PR #44587.

#58196 Investigating P3 comp/agent

[Feature]: Verification enforcement -- verify-before-claim guidance + unsupported-completion-claim detector

Agents routinely claim completion without verifying; the proposal adds verify-before-claim system-prompt guidance plus a log-only detector for unsupported completion claims, already implemented per PR #54576.

#58195 Investigating P3 comp/agent

[Feature]: Per-slot reasoning_effort for MoA reference advisors

MoA reference models spend most of their latency on private reasoning the aggregator never sees, with no way to reduce it independently; the proposal adds a per-slot reasoning_effort option, already implemented per PR #57043.

#58194 Investigating P3 comp/gateway

[Feature]: Excel sidecar -- chat with Hermes inside a live workbook

There is no way to use the assistant directly against a live Excel workbook; the proposal adds an Office.js sidecar with strict tool containment, already implemented per PR #44356.

#58193 Investigating P3 comp/agent

[Feature]: Attach an open desktop window so the agent can see and control it

Operating a desktop app currently requires describing the window and controls in text; the proposal adds a way to attach a live window so the agent can see and control it via a UI-automation sidecar, already implemented per PR #53852.

#58192 Investigating P3 comp/agent

[Feature]: Keep local-backend prompt prefixes warm to eliminate cold-session prefill

Local llama.cpp/vLLM-style servers pay a full prefill cost for every new session because the shared prompt prefix is not kept warm; the proposal adds an opt-in gateway watcher that periodically replays a minimal request, and the author notes it is already implemented in PR #57019.

#58188 Investigating P3 comp/gateway

BlueBubbles needs product-level delivery health checks, not just gateway/process liveness

BlueBubbles/iMessage can silently stop delivering inbound messages to Hermes even while process and port signals still look healthy; the issue proposes a real product-level health check instead of relying on weak liveness signals.

#58170 Investigating P3 comp/tui

Feature request: Mouse support + auto-copy selection (Claude Code parity)

The request asks for Claude Code-style mouse support (cursor placement, expandable sections, scrolling) and auto-copy of selected text to the clipboard in the Hermes CLI/TUI.

#58161 Investigating P3 comp/desktop

[Feature]: I hope the desktop app adds a feature to adjust the interface font size.

User requests a font size adjustment option for the Hermes desktop app to improve readability for users with poor eyesight.

#58151 Investigating P3 comp/cli

[Feature]: CLI commands and agent-assisted session archiving/organization

A power user with hundreds of accumulated sessions requests CLI commands and agent-assisted tools to archive, organize, tag, and export sessions, since none currently exist.

School diagnosis & fix
#58134 Investigating P3 comp/desktop

Enhanced pet display with real-time execution status (inspired by Petdex Desktop)

Requests enhancing the desktop/TUI pet display with a floating window, real-time execution status (current tool, progress, token usage), cross-agent reactivity, and more states, inspired by Petdex Desktop.

#58132 Investigating P3 comp/cli

feat: 会话添加重要性/优先级标记 (Session Importance)

### 问题描述 Hermes 目前没有给会话（session）标记重要性或优先级的功能 ### 建议方案 给每条会话增加一个「重要性/优先级」属性 **1. 设置重要性** - CLI：`her

School diagnosis & fix
#58130 Investigating P3 comp/desktop

Feature Request: Desktop audio playback pill — volume slider, play/pause, speed toggle

Requests adding volume, pause/resume, and playback speed controls to the desktop app's audio/TTS playback pill, which currently only offers a status label and a stop button.

#58125 Investigating P3 comp/cli

Feature Request: Live kanban task status in TUI/CLI

Requests live status indicators (spinners, watch mode, heartbeat display) for kanban tasks in the TUI/CLI, since current output is static and requires manual re-running to see progress.

School diagnosis & fix
#57748 Investigating P3 comp/gateway

[Feature]: Stream live context-window usage to the desktop/TUI status bar while a turn is running

The context-window usage gauge in the desktop/TUI status bar only refreshes on message.complete, so it stays frozen during long agentic turns even though the underlying data updates after every internal API call.

#57742 Investigating P3 comp/dashboard

[Feature]: Directory Download / ZIP Export in Dashboard Files Page

The Files page in the Hermes web dashboard only supports downloading individual files; there is no way to download an entire directory tree at once, which is impractical for backups with thousands of files.

#57738 Investigating P3 comp/desktop

Hermes Desktop: Docker execution backend toggle has no status, health check, or setup guidance

When the Docker execution backend is selected in Hermes Desktop, there is no status indicator, no test-connection action, and no setup guidance if Docker Desktop is not installed or running, resulting in silent generic failures.

#57731 Investigating P3 comp/desktop

[Feature]: Proper update changelog screen

The desktop app's update screen shows only a few recent updates plus a count of remaining ones; the reporter requests a scrollable page listing all accumulated updates.

#56655 Investigating P3 comp/agent

Feature: task-aware per-turn model routing via a pre_llm_call model override

The reporter requests a way for a plugin to choose the model per turn based on the incoming message, noting that existing hooks like pre_llm_call cannot currently change the model used for a turn.

#49730 Investigating P3 comp/gateway

Cross-platform conversation continuity (shared session context across Telegram, iMessage, CLI, etc.)

Requests a mechanism to carry conversation context across platforms (Telegram, iMessage, CLI, Discord), since currently only persistent memory carries over and conversation history stays isolated per platform.

#47349 Workaround P2 comp/agent

Feature: Configurable Memory Backends — disable memory.md, use honcho/fact_store only

This feature request proposes renaming memory.md to rules.md and adding configurable memory backends, since injecting all memory entries into every turn currently mixes always-needed rules with queryable facts and wastes tokens.

#46169 Investigating P3 comp/tui

Desktop should support Ctrl+F/Cmd+F find across chat and UI editors

The request asks for a browser-style Ctrl+F/Cmd+F find UI in Hermes Desktop that can search chat transcripts and editor/settings panels with match highlighting and navigation.

#44428 Investigating P3 comp/gateway

[Feature]: Support Telegram Bot API 10.1 Rich Messages and rich draft streaming

This issue proposes adding support for Telegram Bot API 10.1's new Rich Messages (headings, lists, tables, LaTeX, etc.) to the Telegram gateway, replacing the current plain/MarkdownV2 message streaming approach.

#44300 Investigating P3

[Feature]: Desktop terminal pane should auto-follow agent backend

The desktop app's embedded terminal pane always spawns PowerShell regardless of the agent's configured terminal backend (e.g. WSL2), so it doesn't reflect the agent's actual execution environment; the reporter proposes auto-detecting and matching the backend.

#43397 Investigating P3 comp/gateway

Gateway should preserve forwarded-message metadata for Telegram and other platforms

The gateway currently discards Telegram's native forwarded-message metadata (forward_origin, etc.), so the agent cannot distinguish typed text from forwarded messages; the issue proposes a platform-neutral annotation layer to preserve it.

#41222 Investigating P3 comp/cron

Feature Request: Integrate Kanban Board into Desktop App

The Kanban board is currently separate from the Desktop app, requiring users to run CLI commands and manually copy task IDs; the issue proposes integrating it via a sidebar or in-chat /kanban commands.

#40399 Fixed P3

[Feature]: Hermes Desktop — styling, theming, and font customization

This closed issue proposed a comprehensive styling system for Hermes Desktop covering text formatting, font customization, and a dark/light theme system with custom theme support.

#40297 Investigating P3 comp/desktop

Desktop: make workspace selectable per session, not only launch-time cwd

This issue reports that Hermes Desktop only supports setting a workspace directory at launch time via --cwd, making it inconvenient to work across multiple projects, and proposes per-session workspace selection inside the app.

#40166 Investigating P3 comp/desktop

[Desktop app]: add font size / zoom control

The Hermes desktop app on macOS has no font size or zoom control, standard macOS zoom shortcuts and trackpad pinch don't work, and there is no config.yaml option; the issue requests adding one.

#39691 Investigating P3 comp/agent

feat(compression): integrate headroom-ai for tool output compression

This issue proposes integrating the open-source headroom-ai library to compress individual tool outputs before they enter context, addressing known issues with Hermes's existing conversation-level compression system.

#38975 Investigating P3 comp/tui

Desktop UI: add full custom OpenAI-compatible provider setup

Hermes Desktop only has partial UI support for custom OpenAI-compatible providers, forcing users to overload another provider's fields or manually edit config files; the issue requests a proper Custom provider row needing only a Base URL and API key.

#38969 Investigating P3 comp/tui

Desktop: Add text scaling and configurable text width / container margins

This issue reports that Hermes Desktop's fixed-width chat container is impractical on both large monitors and small screens, and requests configurable text scaling and container width settings.

#38952 Fixed P3 comp/tools

feat: Make Mixture-of-Agents (moa) fully configurable via config.yaml for cost/quality tuning

Feature request (now closed as fixed) to make the Mixture-of-Agents (moa) toolset configurable via a config.yaml moa: section, plus session-level commands to toggle MoA routing on demand.

#38671 Investigating P3

Desktop Remote Gateway should support a true remote workspace/file browser

In Hermes Desktop's remote gateway mode, the file browser reads the local Electron filesystem while chat/session execution runs on the remote backend, creating a split-brain workspace that isn't truly remote-aware.

#38602 Investigating P3 comp/desktop

[Feature]: Desktop Client-Only Installation

This issue requests a first-class client-only install mode for Hermes Desktop that connects to an existing remote Hermes instance, since today the app always attempts to bootstrap a local runtime on first launch unless one is found.

#38519 Investigating P3

[Feature]: Hermes Desktop frontend install only

User requests a frontend-only installation option for Hermes Desktop, to connect to an agent already running on a different machine without installing the agent locally.

#37713 Investigating P3 comp/gateway

Desktop Remote gateway should support switching active Hermes profile from the UI

Hermes Desktop's Profiles page can list profiles but cannot actually switch which profile a connected remote backend uses; the issue requests a first-class profile switcher in the Desktop UI.

#36970 Investigating P3 comp/desktop

Desktop: add first-class remote-client onboarding for existing Hermes instances

This issue requests a first-class remote-client onboarding flow in Hermes Desktop, so users with an existing Hermes backend can connect directly instead of being funneled into the local bootstrap/install flow.

#35750 Investigating P3 comp/cli

[Feature]: Real-time Voice Conversation Mode (voice-in, voice-out, low-latency)

This issue requests a real-time voice conversation mode for Hermes Agent's terminal/TUI, including microphone input, streaming speech-to-text, low-latency responses, and streaming text-to-speech output.

School diagnosis & fix
#34352 Investigating P3 comp/agent

Solving the Multi-Tenant Hermes Problem

This issue reports that memory operations bypass Hermes's hook system, making tenant isolation impossible without forking core, and proposes upstreaming a fix along with an open-source project called Hermes Swarm Map.

#33223 Investigating P3 comp/agent

[Discussion] Why was smart_model_routing removed? Request to restore

This is a discussion asking why the smart_model_routing feature, which auto-routed simple short turns to a cheaper model, was removed entirely rather than kept as opt-in, and requesting it be restored.

#32925 Investigating P3 tool/skills

Feature Request: Integrate Microsoft SkillOpt for Self-Evolving Agent Skills

This issue proposes exploring integration of Microsoft's SkillOpt, a text-space optimizer for training reusable agent skills via trajectory-driven edits, to enable automated skill optimization in Hermes without modifying model weights.

School diagnosis & fix
#30649 Investigating P3 comp/agent

[Feature]: Proton Pass AI Access Tokens support (secret source backend)

Hermes currently supports Bitwarden Secrets Manager as an external secret source; the issue proposes adding Proton Pass's new AI Access Tokens as an additional secret source backend, given its read-only vault access, expiration controls, and audit logs.

#29530 Workaround P2 comp/cli

Profiled workers need a shared auth home separate from isolated HERMES_HOME

Profiled Hermes workers using isolated HERMES_HOME directories can end up with split-brain OAuth state for rotating/single-use refresh tokens; the reporter proposes a separate shared HERMES_AUTH_HOME for auth storage.

School diagnosis & fix
#29299 Investigating P3 tool/mcp

[Feature]: Need Https Oauth callback url ability

Feature request to support an HTTPS OAuth redirect_uri override for MCP, since some OAuth providers (e.g. Salesforce) reject the current localhost HTTP callback format.

School diagnosis & fix
#25267 Investigating P3 comp/plugins

[Feature]: Claude Agent SDK model provider with subscription OAuth (Codex-style)

This issue requests a Claude subscription OAuth model provider for Hermes (similar to the existing Codex OAuth flow), so Claude subscribers can use their monthly Agent SDK credit instead of paying separately for a Developer Platform API key.

#24039 Workaround P2 comp/agent

Auxiliary fallback chain should reuse fallback_providers, not maintain a separate hardcoded list

Hermes maintains two separate fallback chains that are unaware of each other: the user-configured fallback_providers used by the main agent, and a separate hardcoded fallback list (including paid models) used by auxiliary tasks like compression, vision, and title generation.

#23717 Investigating P3 comp/agent

RFC: Pluggable SessionDB Provider — PostgreSQL, MySQL, and Beyond

This RFC proposes a pluggable SessionDB backend (e.g. PostgreSQL, MySQL) to replace Hermes's current shared SQLite state.db, which is prone to lock contention and corruption when multiple processes run concurrently with hot updates.

#22791 Investigating P3

[Feature]: Add Infisical as an External Vault backend (sub-issue of #3630)

The issue requests adding Infisical as an External Vault backend, since Hermes currently has no native way to consume secrets from Infisical's recommended multi-project layout, forcing users into hard-coded shell wrappers or a single flat project.

#21587 Investigating P3 comp/gateway

Feature: Telegram Guest Bots, Bot-to-Bot, Stickers and Chat Automation

This feature request proposes adopting several of Telegram's May 2026 bot platform updates, notably Guest AI Bots (usable via @mention without joining a chat) and Bot-to-Bot communication, to support Hermes multi-agent workflows.

#20859 Investigating P3

[Feature]: Support for Mistral as LLM provider

This issue requests native support for Mistral as an LLM provider in Hermes-Agent, since it currently can only be used via the less convenient custom provider option.

#20510 Investigating P3

Feature Request: Cloud Sync for All Hermes Configurations Across Devices

Hermes stores all config, profiles, skills, memory, and sessions locally with no built-in cross-device sync, requiring manual export/import today; the issue proposes a built-in cloud sync feature.

#20318 Investigating P3 comp/gateway

[Feature]: TrueConf gateway / messaging platform support

The issue requests adding TrueConf as a supported Hermes messaging gateway, similar to existing Slack/Discord/Telegram support, noting a ready-made python-trueconf-bot library is available.

#20249 Investigating P3 comp/agent

Model Presets: per-turn expert-on-demand model escalation

This feature proposes a model_presets config section that lets users temporarily escalate to a named expert model for a single turn and then automatically snap back to the default model, instead of manually switching models back and forth.

#18918 Fixed P2 comp/gateway

Slack: render Markdown pipe tables as Block Kit tables

Slack's mrkdwn does not render GitHub-Flavored Markdown pipe tables, so Hermes responses containing tables displayed as literal pipe text in Slack; the issue is now closed.

#18733 Investigating P3 comp/agent

Per-model or per-provider compression threshold overrides

Feature request to allow per-model or per-provider overrides of compression.threshold in config.yaml, since a single global threshold is ineffective for large-context models and too aggressive for small-context ones.

#18715 Investigating P3 comp/agent

Support remote Hermes agent with local tool execution

This issue requests a split-runtime mode where a remote Hermes Agent provides skills, memory, and model reasoning, while tool execution (terminal, files, browser, local MCP servers) happens on the local client machine instead of on the remote host.

#18080 Workaround P2 comp/dashboard

[Feature]: Improved Themes for Dashboard - currently hard to read

This issue requests new dashboard themes with more standard fonts, sizes, weights, and contrast, since the existing themes (Midnight, Ember, Mono, Cyberpunk, Rose) are reported as hard to read, citing Linear as a reference for clarity.

#16702 Investigating P3 comp/gateway

Feature request: Auto-start dashboard with gateway systemd service

User reports that the Hermes dashboard runs as a separate process and does not restart automatically after WSL reboots or gateway crashes, and proposes a systemd service to auto-start it after the gateway.

#16640 Investigating P3

[Feature]: Docker stable releases

The nousresearch/hermes-agent:latest Docker image updates every few hours, which is too frequent for some users; the issue requests a stable release tag/channel instead of manually pinning specific version tags.

#16102 Fixed P3 comp/cli

RFC: review the Kanban — multi-profile collaboration board (PR #16100)

This RFC tracks the implemented Kanban board PR (#16100): a long-lived daemon replacing the cron-driven dispatcher, a drag-and-drop dashboard plugin, run history and worker log panels, live WebSocket updates, and multiple audit/review passes with concurrency and fuzz testing. Issue is closed as fixed, with further discussion moved to the PR.

School diagnosis & fix
#15602 Investigating P3 tool/skills

google-workspace skill: add multi-account support

The google-workspace skill only supports a single Google account due to a fixed token file path, forcing users with multiple accounts to route the second account through an external tool; the issue proposes an --account flag with per-account token files.

School diagnosis & fix
#15445 Investigating P3 comp/tools

feat(browser): add Obscura as a browser provider (Rust headless browser with CDP)

This issue proposes adding Obscura, a Rust-based headless browser with a CDP-compatible WebSocket server, as a browser provider option for Hermes, citing lower memory/binary size and faster startup compared to Chrome/Chromium.

#15311 Investigating P3 comp/gateway

Add generic action buttons / inline keyboard support for messaging platforms

Feature request for a generic, platform-agnostic API for attaching interactive action buttons (especially Telegram inline keyboards) to any message, instead of hardcoding button support per specific built-in feature.

#14853 Investigating P3 comp/gateway

[Feature]: Multi-Agent Discord channel collaboration — history injection + cascade prevention

Feature proposal to add Discord channel history injection and cascade-prevention/mention-gating logic for multi-agent shared-channel setups, based on a working patch the reporter built for their own 3-agent deployment.

#13484 Investigating P3 comp/agent

Feature: native Google Cloud Vertex AI provider support

Hermes has no working auth path for the Google Cloud Vertex AI provider, causing silent failures; the reporter built a standalone proxy handling service-account auth and proposes upstreaming it via the existing custom_providers mechanism.

#13181 Investigating P3 comp/agent

[Feature]: Easy support for adding OpenCode Go models into Hermes Agent

Integrating new model backends like OpenCode Go into Hermes Agent currently requires digging into internal code; the issue proposes a config-based model registration system with a standardized adapter interface.

#13065 Fixed P3 comp/agent

Feature: First-class native vision support for vision-capable main models (with reference implementation + bug findings)

This feature request notes that Hermes routes all image analysis through an auxiliary vision model even when the main model is natively vision-capable, and provides a reference implementation (patching 4 files) for native vision bypass along with several related pipeline bugs discovered in the process.

#12639 Investigating P3

[Feature]: Support for Native Google / Vertex AI Provider (Bypass OpenRouter 402 Errors & Rate Limits)

User reports frequent HTTP 402 and rate-limit errors from OpenRouter when using the google/gemini-3.1-pro-preview model, and requests a native Google/Vertex AI provider option to bypass OpenRouter entirely.

#12572 Investigating P3

[Feature]: Provide a lightweight Docker image without Playwright/Chromium

This issue reports that the official Docker image is about 2.4GB largely due to Playwright/Chromium dependencies, and requests a lightweight image tag that excludes browser automation for users who don't need it.

#12240 Fixed

[Feature]: Add Yuanbao (元宝) as a new gateway platform

Feature request to add Tencent's Yuanbao AI-social platform (with its 'Pai' group chats) as a new Hermes gateway, complementing existing support for Telegram, Discord, Slack, WhatsApp, and several Chinese enterprise platforms.

#12238 Investigating P3 comp/cli

Feature Request: Built-in Automatic Backup & Version Control for Agent Data (~/.hermes/)

This issue requests a built-in automatic backup and version control mechanism for all Hermes agent data stored in ~/.hermes/, including a proposed `hermes backup` CLI command for immediate or scheduled snapshots.

School diagnosis & fix
#11911 Investigating P3 comp/desktop

[Feature Request] Native Mobile App (iOS & Android) with Voice Calling

This issue requests an official native mobile app for iOS and Android supporting real-time voice calling with the Hermes AI assistant.

#11692 Investigating P3 comp/agent

Receipts for self-improving agents: proving which skill version produced which output

A discussion-style feature request raises governance concerns about Hermes' self-modifying skills: since skills are created and improved automatically, there is no built-in way to prove which version of a skill produced a given output or under what policy it ran.

#11420 Fixed P3 comp/agent

Add MiniMax as vision backend in auxiliary_client.py (_try_minimax)

Configuring AUXILIARY_VISION_PROVIDER=minimax silently fails the vision_analyze tool because _resolve_strict_vision_backend() has no MiniMax branch, even though MiniMax offers multimodal vision endpoints.

#10644 Investigating P3 comp/tools

Feature: Add Brave Search as a native web search backend

This issue proposes adding Brave Search API as a first-class native web search backend in hermes-agent, alongside existing providers like Firecrawl, Parallel, Tavily, and Exa.

#10567 Investigating P3 comp/cli

Add --host and CORS config for hermes dashboard to enable Tailscale/VPN access

The hermes dashboard's CORS regex is hardcoded to localhost/127.0.0.1, so even with --host 0.0.0.0 the frontend loads but API calls are blocked, preventing remote access over VPNs like Tailscale; the issue requests configurable host/port and CORS origins.

School diagnosis & fix
#10359 Investigating P3

[Feature]: native windows support

User requests native Windows support for Hermes Agent instead of running via WSL2. This issue is labeled as a duplicate.

#10143 Investigating P3 comp/gateway

[Feature] Topic-to-Profile routing: dispatch messages to different profiles based on Telegram topic/thread

Feature request to let a single Telegram bot route messages from different forum topics/threads to different Hermes profiles (each with its own model, skills, and system prompt), instead of requiring one bot token and gateway process per profile.

#9514 Investigating P3 comp/agent

Feature: Single-Daemon Multi-Agent with Per-Topic Workspace & Memory Isolation

This feature proposes a single-daemon architecture, inspired by OpenClaw, letting multiple Hermes agents share one gateway process with isolated per-agent workspaces and memory, routed by session key including per-topic isolation in group chats.

#9459 Investigating P3 comp/agent

feat(delegation): agent profiles for delegate_task — custom orchestration harness support

This issue proposes allowing delegate_task to spawn subagents from named agent profiles defined in config.yaml, so users can customize system prompts, models, and tool scoping per subagent role without modifying Hermes core.

#9269 Investigating

feat: Tailscale serve integration for secure remote access

This issue proposes optional Tailscale serve integration so users can securely expose the Hermes API server over their Tailnet without public ports or reverse proxies.

#9056 Investigating

[Feature]: Add a Nix Home Manager module for user-level installation

This issue requests a Nix Home Manager module for declarative, user-level installation of Hermes Agent, since current Nix support is only at the NixOS system level.

#8552 Workaround P2 comp/gateway

Slack platform: use Block Kit markdown block type instead of legacy mrkdwn

This issue reports that Hermes's Slack adapter uses the legacy mrkdwn format, which doesn't support markdown tables, and proposes switching to Slack's newer Block Kit markdown block type.

#8461 Fixed P3

Feature Request: Add glm-5.1 model to Z.AI Coding Plan

This issue requested adding the glm-5.1 model to the Z.AI provider's curated model list for Coding Plan users; the issue is now closed.

#8457 Investigating P3 comp/agent

Feature: Persistent Session Memory with Cross-Session Search & Auto-Compression

This feature request proposes a persistent, searchable, auto-compressing "Vault" style session memory system to replace Hermes's current transient memory, which is lost whenever a session ends or the gateway restarts.

#8366 Investigating P3 comp/gateway

Feature: Cross-Platform Session Handoff (CLI ↔ Telegram ↔ iMessage)

Hermes sessions are isolated per platform (CLI, Telegram, iMessage) with no continuity, so work cannot be resumed across platforms; the issue proposes a built-in handoff mechanism to carry state between them.

#8287 Investigating P3 comp/gateway

Support multiple Telegram bots connecting to the same agent (same gateway, different sessions) using the same Telegram account.

Hermes currently only supports parallel background tasks via /background, which lacks mid-task interaction and runs sequentially; the issue requests allowing multiple Telegram bots on one account to connect to the same agent with independent sessions.

#7517 Investigating P3 comp/agent

Feature Request: Native Multi-Agent Support

This issue requests native multi-agent support in Hermes, allowing a single gateway process to serve multiple named agents with isolated sessions, personas, memory, and tool configurations, instead of today's single global agent identity.

#6839 Investigating P3 comp/agent

Feature: Lazy Tool Schema Loading — Two-Pass Tool Injection to Reduce Token Overhead

This issue proposes a two-pass lazy tool schema loading scheme (abbreviated tool list first, full schema only when a tool is actually selected) to cut the per-call token overhead of injecting all enabled tool schemas.

#6323 Fixed P3 comp/plugins

add mempalace for external memory support

This issue proposes adding a mempalace module to Hermes for structured external memory, providing persistent, queryable memory beyond the context window with semantic retrieval. This issue is closed.

#6296 Investigating

[Feature]: Mattermost adapter should register COMMAND_REGISTRY as custom slash commands with autocomplete

The Mattermost mobile client intercepts any message starting with / as a native slash command, breaking Hermes's slash commands including the critical /approve flow; the issue requests registering COMMAND_REGISTRY as native Mattermost slash commands like other adapters already do.

#5997 Investigating

[Feature]: LLM model switch by skill

This issue requests letting individual skills specify or be assigned a cheaper LLM model instead of always inheriting the main agent's globally configured model.

#5941 Investigating P3 tool/web

[Feature]: Add Searxng as a default web search provider (alongside firecrawl, tavily, etc)

This issue proposes adding Searxng as an additional native web search backend in web_tools.py, alongside existing providers like Firecrawl and Tavily.

#5826 Investigating P3 comp/gateway

[Feature]: Linear platform adapter for the gateway

This feature request proposes adding a Linear platform adapter to the gateway, allowing users to interact with the agent directly via @mentions in Linear issue comments instead of switching to Slack, Telegram, or Discord.

#5712 Workaround P2 comp/gateway

[Feature]: True Autonomy - Automatically Inject Cron Results into Live Gateway Chat Sessions

Hermes cron job results delivered to the origin channel are visible only to the human and not mirrored into the live gateway session's context, so the main agent has no memory of completed scheduled tasks unless a human manually points it to a filesystem handoff file.

#5533 Investigating P3 comp/agent

feat(dreaming): introduce stable Dreaming reflection mode across CLI and gateway

This issue documents making Dreaming a first-class, reliable reflection feature across both the CLI and gateway (including Telegram), with tool usage disabled, correct runtime/provider config propagation, and error surfacing instead of silent failure.

#5528 Investigating P3 comp/agent

[Feature]: configurable approval-locked command patterns for dangerous/disruptive local actions

The set of commands requiring dangerous-action approval is currently hard-coded in tools/approval.py, so users cannot mark installation-specific or operationally-disruptive commands as approval-required without patching the source; the issue requests a configurable mechanism.

#5346 Fixed P3 comp/cli

[UX] Add Shift+Enter newline support in CLI input

This issue proposes adding Shift+Enter as an additional newline keybinding in the Hermes CLI/TUI input, alongside the existing Alt+Enter and Ctrl+J bindings. This issue is closed.

School diagnosis & fix
#5257 Investigating P3 comp/agent

feat: Generalized ACP client for multi-agent CLI orchestration

This issue proposes generalizing Hermes's ACP client beyond the current Copilot-specific implementation to orchestrate 14 ACP-compatible coding agents (Claude Code, Codex CLI, Gemini CLI, etc.) using official vendor-maintained ACP adapters.

#5143 Investigating P3 comp/gateway

[Feature] Multi-Role Auto-Routing via Gateway Hooks

This issue proposes Multi-Role Auto-Routing, letting users define named roles with dedicated sessions and system prompts, auto-routed by a lightweight classifier, to avoid context pollution from mixing unrelated topics in one shared session.

#4983 Fixed

[Feature]: Native Google GenAI Provider for Gemini Models

Feature request for a native Google GenAI provider so Gemini models can be used directly instead of through a fragile OpenAI-compatibility translation layer that causes unstable tool calling and dropped streaming tokens. Issue is closed as fixed.

#4656 Investigating P3 comp/agent

[Feature]: credential proxy daemon — zero-knowledge HTTP/HTTPS broker for agent credentials

This feature proposes a credential proxy daemon that intercepts HTTP/HTTPS at the transport layer so the real credential value never exists anywhere the agent process can read it, addressing a gap left by existing env-scoping and PID-isolation mitigations.

#4505 Investigating P3 comp/agent

Optimize Ollama Integration: Native /api/chat vs OpenAI-Compatible Endpoint

Feature proposal to replace the OpenAI-compatible endpoint with Ollama's native /api/chat endpoint, claiming true delta streaming, longer timeouts, full native parameter support, and ~15-20% lower latency, with a proposed 581-line adapter implementation.

#4431 Investigating P3 comp/gateway

[Feature]: Per-topic/per-chat configuration overrides (topic_configs) for gateway sessions

This issue reports that all Hermes gateway sessions across multiple Telegram groups or Discord servers currently share the same personality, prompt, and working directory, and proposes a topic_configs config section for per-chat overrides.

#3913 Fixed

[Feature]: Docker images for linux/arm64 missing

User requests linux/arm64 Docker images for running Hermes Agent on Apple Silicon Macs, since only linux/amd64 images are currently available. This issue is closed.

#3863 Fixed

[Feature]: Native AWS Bedrock provider support

This issue requests native direct integration with AWS Bedrock using the standard AWS SDK credential chain, since accessing Bedrock only via OpenRouter today adds latency, cost markup, and loses AWS account-native credentials and enterprise features. This issue is closed.

#3725 Investigating P3 comp/gateway

[Feature]: Rocket Chat support

The issue requests adding Rocket.Chat as a supported Hermes messaging gateway channel.

#3326 Investigating P3 comp/cli

feat(cli): add --output-format json flag for structured query output

hermes chat -q currently only produces plain text output for programmatic use, requiring manual parsing of token counts, cost, and session metadata; the issue requests a --output-format json flag, with an associated PR already filed.

School diagnosis & fix
#2988 Investigating

[Feature]: XMPP chat support with OMEMO encryption

This issue proposes adding support for the XMPP protocol with OMEMO encryption as a self-hostable, secure real-time chat option, since current support is limited to proprietary platforms like WhatsApp and Signal.

#2512 Fixed

[Feature]: Native Windows Support

This closed issue requested native Windows support for Hermes Agent.

#1955 Fixed P2 comp/gateway

feat: per-channel model and system prompt overrides for gateway platforms

This feature proposes a channel_overrides config section so each Discord/Telegram channel can specify its own model, provider, and system prompt instead of sharing one global model and prompt across all channels.

#1855 Investigating P3 tool/terminal

feat(terminal): Multi-backend terminal — local + N named remotes with persistent shell

This issue reports that Hermes Agent only supports one terminal backend at a time, forcing users into fragile one-shot SSH commands for multi-machine workflows, and proposes supporting local execution plus multiple named persistent remote backends simultaneously.

#1050 Fixed

[Feature]: Homebrew formula for hermes-agent

This issue requests a Homebrew formula for hermes-agent; the reporter submitted one via a PR to Homebrew/homebrew-core. This issue is closed.

#850 Fixed

[Feature]: Full docker deployment

This issue requests a fully dockerized deployment mode for hermes-agent, where the main task also runs in a container and communicates with the dockerized agent sandbox over http/ws. This issue is closed.

#569 Investigating

Feature: Agent Client Protocol (ACP) Server Mode — Run Hermes in Zed, JetBrains, Neovim, Toad & Any ACP-Compatible Editor

This issue proposes implementing the agent side of the Agent Client Protocol (ACP) in Hermes so it can run inside any ACP-compatible editor or client, such as Zed, JetBrains IDEs, Neovim, and Toad.

#527 Workaround P2 comp/gateway

Feature: Gateway Permission Tiers — Role-Based Access Control (Owner/Admin/User/Guest) for Messenger Platforms

This issue reports that Hermes Agent's gateway authorization is currently all-or-nothing, and proposes a tiered permission system (Owner/Admin/User/Guest) for messaging platforms.

#514 Investigating P3 comp/agent

Feature: A2A (Agent-to-Agent) Protocol Support — Remote Agent Discovery, Communication & Interoperability

This issue proposes adding support for Google's A2A (Agent-to-Agent) protocol to Hermes, allowing it to discover and call remote agents built on other frameworks and expose itself as an A2A-discoverable agent.

#501 Fixed

Feature: Web UI Gateway — Local Browser-Based Interface with Streaming, Artifacts & Rich Rendering

This closed issue proposed adding a local browser-based Web UI gateway for Hermes Agent with streaming, artifacts, and rich rendering, based on research into 30+ existing AI interface projects.

#487 Fixed

Feature: Cryptographic Audit Trail — SHA-256 Hash-Chained Action Log for Tamper-Proof Agent Accountability (inspired by OpenFang)

This closed issue proposed adding a cryptographic, SHA-256 hash-chained audit trail for agent actions in Hermes, inspired by the OpenFang project's tamper-evident logging design.

#410 Fixed

Feature: Secure Secrets Management Tool — API Key Ingestion, Scoped Access, Redaction, and Skill Requirements

This umbrella issue described the lack of secure secret handling in Hermes: secrets stored as plaintext env vars fully exposed to subprocesses and file tools, with no mechanism for skills to safely request credentials; the issue is now closed.

#344 Fixed

Feature: Multi-Agent Architecture — Orchestration, Cooperation, Specialized Roles & Resilient Workflows

This umbrella issue proposes evolving Hermes from a single agent with isolated sub-agent delegation into a true multi-agent architecture with specialized roles, dependency-aware workflows, inter-agent cooperation, and resilient execution. This issue is closed.

#157 Fixed P3 comp/agent

Feature Request: User-Configurable Multi-Model Routing with Capability Categories and Evaluation Feedback

The issue proposed letting users assign multiple LLMs to capability categories (speed, cost, reasoning depth, etc.) so tools could dynamically request a model based on declared needs rather than a single hard-coded model; the issue is now closed.

#73 Fixed P3 comp/gateway

Matrix Protocol Support for Messaging Gateway

The issue requested native Matrix protocol support as a Hermes messaging gateway so users could interact via any Matrix client on self-hosted homeservers; the issue is now closed.

Other categories

Back to curated issues . School solution cards . Troubleshooting